Pandas column contains a series of urls. I'd like to extract a substring from the url.
MRE code below.
s = pd.Series(['https://url-location/img/xxxyyy_image1.png'])
s.apply(lambda x: x[x.find("/")+1:st.find("_")])
I'd like to extract xxxyyy and store them into a new column.
You can use
>>> s.str.extract(r'.*/([^_]+)')
0
0 xxxyyy
See the regex demo. Details:
.* - zero or more chars other than line break chars as many as possible/ - a slash([^_]+) - Capturing group 1 (the value captured into this group will be the actual return value of Series.str.extract): one or more chars other than _ char.
/img and know to look at the one _ in the substring?.* is a greedily quantified pattern, it grabs the whole string at first. The engine starts backtracking then, trying to match some text with the subsequent patterns. So the / char found is the last / char that is followed by one or more chars other than _. The [^_] is a negated character class, it matches any char other than a _ char, so it cannot match across more _s, it will stop before the first _ or end of string. Here is my YT video about backtracking in regex.Also possible:
s.str.split('/').str[-1].str.split('_').str[0]
# Out[224]: xxxyyy
This works, because .str allows for the slice annotation.
So .str[-1] will provide the last element after the split for example.
